Base editing treats spinal muscular atrophy in mice

A one-time genome-editing treatment restores motor function and extends lifespan in an animal model of a neuromuscular disease that is the leading genetic cause of infant mortality.

Fluorescence image of the lumbar spinal cord in a mouse model of spinal muscular atrophy
Credit: Dan Wang and Ailing Du
Fluorescence image of the lumbar spinal cord in a mouse model of spinal muscular atrophy, showing cell nuclei (blue), astrocytes (red), and AAVs (green), viral packaging used to deliver the base-editing treatment.

Researchers have used a gene-editing technique called base editing to restore motor function to near-normal levels in a mouse model of spinal muscular atrophy (SMA) — a disease that leads to paralysis and, in its most severe form, death before the age of two in humans.

SMA occurs when cells in the spinal cord that control muscle movement, called motor neurons, die. Patients with the disorder have low levels of the SMN protein because they either have mutations in the SMN1 gene or are missing the gene entirely.  

The scientists, from the Broad Institute of MIT and Harvard, The Ohio State University Wexner Medical Center, and the University of Massachusetts Chan Medical School and led by David Liu, a core institute member at Broad, focused their strategy on editing a nearly identical gene, SMN2, because they wanted to develop a treatment that works for all patients. People with SMA can have different SMN1 mutations or lack SMN1, but they all have the SMN2 gene, which also produces the SMN protein. However, that protein is truncated and degrades quickly, resulting in SMN protein deficiency.

In a study published today in Science, the team showed how they used base editing — a gene-editing technique developed by the Liu lab — to replace a single thymine (T) base in SMN2 with cytosine (C), effectively converting it into a functioning copy of SMN1. The strategy restored the SMN protein to normal levels, while also preserving the natural regulatory mechanisms that control the production of the protein.

The scientists tested their approach in a mouse model of spinal muscular atrophy and found that animals that received a single treatment of both the gene-editing system and an FDA-approved SMA drug called nusinersen showed no significant difference in muscle strength, coordination, and physical activity levels compared to healthy mice. The treated mice also lived on average 6.5 times longer than untreated animals with the disease.      

The team compared their gene-editing approach with nusinersen and two other FDA-approved treatments for SMA and found that base editing increased motor function in mice similarly or more than each of the three existing drugs. However, the scientists discovered the most dramatic improvements when they combined the base-editing treatment with nusinersen. The results suggest that base editing might provide a one-time therapy for SMA and improve existing drug treatments.

“The three FDA-approved drugs have revolutionized the treatment of SMA for thousands of patients,” said Liu, who is senior author on the study and also the Richard Merkin Professor and director of the Merkin Institute of Transformative Technologies in Healthcare at the Broad, as well as a Howard Hughes Medical Institute investigator and a professor at Harvard University. “But one of the real promises of precision gene-editing therapies is the possibility that a one-time treatment can provide a therapy for the lifetime of the patient.”

Mandana Arbab, a postdoctoral fellow, and Zaneta Matuszek, a graduate student, both in Liu’s lab, are co-first authors on the study. They collaborated with a team from The Ohio State University Wexner Medical Center led by Arthur Burghes.

“Base editing is a powerful tool to correct genetic diseases,” said Arbab, who was recently appointed as an assistant professor at Harvard Medical School and the Boston Children’s Hospital Translational Neuroscience Center. Her lab will focus on developing gene editing for SMA and other neurological diseases for the clinic. “Most drugs find ways to compensate for what has already gone wrong in a cell, but here we use base editing to stop SMA where it originates, in the DNA.”

Gene fix

Existing treatments for SMA, while effective, require multiple doses, do not fully restore SMN protein levels, subvert the natural regulatory mechanisms that control SMN production, may lose effectiveness over time, or may provoke an immune reaction upon repeated dosing. 

Liu’s team sought to develop a gene-editing approach targeting SMN2, which differs from SMN1 by only one base. The researchers thought that by editing SMN2, they could permanently restore levels of the normal SMN protein while maintaining normal cellular regulation of SMN expression.

Using a combination of machine learning and experimental testing, the scientists evaluated 79 different nuclease-editing and base-editing strategies targeting different parts of the SMN2 gene and homed in on one that fully restored levels of the SMN protein with few editing byproducts and little editing elsewhere in the genome. The team ultimately chose an adenine base editor to convert the T•A base pair in the SMN2 gene to a C•G pair, effectively turning the gene into a healthy copy of the SMN1 gene. The treatment increased SMN protein levels 40-fold, restoring them to levels found in healthy cells.

The team next used a pair of viruses commonly used in gene therapy called adeno associated viruses (AAVs) to deliver the base editor to the central nervous system in a mouse model of the disease. About 43 percent of spinal motor neurons received the AAV-delivered base editor cargo, and among those neurons, 87 percent showed conversion of SMN2 to SMN1. The animals receiving the base editor treatment retained their motor function and lived longer. By fully restoring SMN protein levels while preserving how they are regulated in cells, the team’s strategy might make treatment of SMA safer and more effective than other options in the long term, while eliminating the need for repeated dosing. 

Path to the clinic

The researchers said that the striking improvements they observed in mice treated with both gene editing and the SMA drug nusinersen likely occurred because nusinersen extended the very short window of time when motor neurons can be rescued in the animals, allowing the base editor more time to convert SMN2 to SMN1. In patients, this window of opportunity is much longer — months rather than days in mice — and so Liu is hopeful that the treatment might be even more successful in humans. He added that since SMA patients would likely be taking an existing SMA drug before receiving any new experimental treatment, a combination therapy is a promising option for any future clinical trials that test gene editing for SMA.

Liu also said that using prime editing, a versatile gene-editing method that had not yet been developed when his team chose the editing strategies in this study, may further improve gene editing for SMA in the future.

In the meantime, Liu’s lab is developing a simpler single-AAV delivery system that could reduce the dose and simplify their treatment approach for SMA and progeria, a genetic condition that Liu’s lab and their collaborators have successfully treated with base editing in mice. Liu said they hope the improved delivery system could provide a path to clinical trials for these genetic diseases, as he and Arbab are working on for SMA. 

“There's still a huge amount of work to do,” he said. “But achieving a good understanding of how many different gene editing strategies each affect SMA is an important start to our no-stone-unturned approach to developing one-time treatments for SMA and other serious genetic diseases.”


This work was supported in part by the National Institutes of Health, the Bill and Melinda Gates Foundation, the Howard Hughes Medical Institute, and the Friedreich’s Ataxia Research Alliance (FARA).

Paper cited

Arbab M, Matuszek Z et al. Base editing rescue of spinal muscular atrophy in cells and in mice. Science. Online March 30, 2023. DOI:10.1126/science.adg6518